There are many situations in which GPS is either unable to provide the desired level of accuracy or is unavailable. Use of a pseudolite-based reference system for nav- igation can be a means for positioning during these times. While there are advantages in using a pseudolite-based reference system, there are still implementation issues and de ciencies that must be addressed. In many cases a pseudolite system with ground based transmitters has di culty in determining the height of the receiver accurately. This is due to the poor vertical observability inherent in the geometry of the system. A common approach in naval applications for solving the problem of poor vertical observability is to use a height constraint, which is well known when travelling on a surface of water. For a ground-based vehicle, knowledge of the surface topography can be obtained, but cannot be readily used in the same manner as marine cases, since the height is often a varying function of position. This research investigates and develops ve methods of incorporating the known surface topography in a non-linear batch least squares estimation algorithm using carrier-phase measurements from pseudolites. The -oating point carrier-phase ambi- guities are estimated in this process. Real and simulated data sets are used to evaluate the performance of the ve algorithms. In simulation, all methods performed equiv- alently well on a -at surface. When simulating a hill, constraining the solution to lie in a plane tangent to the surface topography appeared to aid the solution with the best knowledge of the terrain. Use of a pseudo-measurement, a commonly used approach, did not provide the best results, and indicates the inadequacy of using this method for pseudolite-based systems. Results using data from a real system on a ground-based vehicle demonstrated sub-decimeter level positioning accuracy in all three dimensions. | Methods For Aiding Height Determination In Pseudolite-based Reference Systems Using Batch Least-squares Estimation by John Harold Robert